Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 1782 in Layer 7

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.1477. Min Act: -0.0000

Data Index: 1215714 (C4 (Web Text))

Max Activating Token Index: 782

Text #1

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.1967. Min Act: -0.0000

Data Index: 125060 (C4 (Web Text))

Max Activating Token Index: 838

Text #2

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.0060. Min Act: -0.0000

Data Index: 988745 (C4 (Web Text))

Max Activating Token Index: 742

Text #3

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.1049. Min Act: -0.0000

Data Index: 945001 (C4 (Web Text))

Max Activating Token Index: 792

Text #4

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.0250. Min Act: -0.0000

Data Index: 1051455 (C4 (Web Text))

Max Activating Token Index: 270

Text #5

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.0737. Min Act: -0.0000

Data Index: 1043398 (C4 (Web Text))

Max Activating Token Index: 552

Text #6

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.0020. Min Act: -0.0000

Data Index: 967045 (C4 (Web Text))

Max Activating Token Index: 971

Text #7

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.0195. Min Act: -0.0000

Data Index: 1265919 (C4 (Web Text))

Max Activating Token Index: 777

Text #8

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.0041. Min Act: -0.0000

Data Index: 104376 (C4 (Web Text))

Max Activating Token Index: 310

Text #9

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.9501. Min Act: -0.0000

Data Index: 627209 (C4 (Web Text))

Max Activating Token Index: 844

Text #10

Max Range: 1.1967. Min Range: -1.1967

Max Act: 1.0388. Min Act: -0.0000

Data Index: 1257489 (C4 (Web Text))

Max Activating Token Index: 311

Text #11

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.9398. Min Act: -0.0000

Data Index: 19073 (C4 (Web Text))

Max Activating Token Index: 80

Text #12

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.9122. Min Act: -0.0000

Data Index: 770957 (C4 (Web Text))

Max Activating Token Index: 484

Text #13

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.8869. Min Act: -0.0000

Data Index: 530080 (C4 (Web Text))

Max Activating Token Index: 31

Text #14

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.9932. Min Act: -0.0000

Data Index: 1082217 (C4 (Web Text))

Max Activating Token Index: 601

Text #15

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.9833. Min Act: -0.0000

Data Index: 266182 (C4 (Web Text))

Max Activating Token Index: 594

Text #16

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.8935. Min Act: -0.0000

Data Index: 560112 (C4 (Web Text))

Max Activating Token Index: 204

Text #17

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.8894. Min Act: -0.0000

Data Index: 113782 (C4 (Web Text))

Max Activating Token Index: 318

Text #18

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.8484. Min Act: -0.0000

Data Index: 1216944 (C4 (Web Text))

Max Activating Token Index: 780

Text #19

Max Range: 1.1967. Min Range: -1.1967

Max Act: 0.8427. Min Act: -0.0000

Data Index: 591197 (C4 (Web Text))

Max Activating Token Index: 958